Re: [PATCH 4/9] tests: remove test-gdbstub.py

[Ilya Leoshkevich <iii@linux.ibm.com>]

On Wed, 2023-08-16 at 13:33 +0100, Alex Bennée wrote:
> 
> Ilya Leoshkevich <iii@linux.ibm.com> writes:
> 
> > On Tue, 2023-08-15 at 15:51 +0100, Alex Bennée wrote:
> > > This isn't directly called by our CI and because it doesn't run
> > > via
> > > our run-test.py script does things slightly differently. Lets
> > > remove
> > > it as we have plenty of working in-tree tests now for various
> > > aspects
> > > of gdbstub.
> > > 
> > > Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> > > ---
> > >  tests/guest-debug/test-gdbstub.py | 177 ------------------------
> > > ----
> > > --
> > >  1 file changed, 177 deletions(-)
> > >  delete mode 100644 tests/guest-debug/test-gdbstub.py
> > 
> > There doesn't seem to be a hbreak test elsewhere, but according to
> > a
> > comment in tcg/multiarch/gdbstub/memory.py it would be mapped to a
> > normal break anyway.
> 
> It is for TCG but for other accelerators there will be different
> handling (although I'm fairly sure only x86 and aarch64 are currently
> plumbed to use the CPUs hbreak bits on KVM).
> 
> However this particular script was a very early addition when I was
> testing stuff manually with images I'd built on my system. If we want
> to
> exercise the gdbstub for accelerators it might be better porting the
> test to avocado?

That would be good, yes.

I was always wondering if the TCG sotfmmu tests could be used to test
the other accelerators? At least for s390x there is nothing
TCG-specific there (besides that they try to trigger TCG-specific
problems), and I sometimes run them manually with KVM as a sanity
check.

> 
> > 
> > Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
> 
>